About me

Hi! I am a second-year Computer Science PhD student at Duke University, supervised by Prof. Monica Agrawal. Previously, I received my Bachelor of Science in Computer Science and Applied and Computational Mathematics from University of Southern California, where I was fortunate to work with Prof. Yan Liu, Prof. Jesse Thomason, and Prof. Maja Matarić.

Research: I study how data shapes LLM behavior and how LLMs interpret data, with the broader goal of understanding what makes these systems reliable, interpretable, and robust. My work examines how pretraining data and inference-time inputs influence model reasoning, including how faithfully LLMs acquire probabilistic knowledge from text. My research also spans diverse data (text, time series, and multimodal inputs), including developing forecasting methods and studying model behavior under distribution shift. More recently, I have been investigating how LLMs identify and explain distributional differences between text corpora, and whether different agent harnesses affect performance in scientific discovery. I am also interested in human–agent interaction, particularly when and how agents should seek human input. My broader application interests lie in AI for healthcare, where generalizability and responsible deployment are especially important.

Feel free to reach out to me if you'd like to discuss research or explore potential collaboration!